Clinical characteristics and outcomes of critically ill women with gynecologic malignancies: a two-database retrospective study using MIMIC-IV v3.1 and eICU-CRD v2.0.

BACKGROUND

Women with gynecologic malignancies may require intensive care unit (ICU) admission for postoperative complications, infection, organ dysfunction, treatment-related toxicity, or advanced cancer. Multicohort ICU outcome data for this population are limited. We described characteristics and in-hospital outcomes using two public critical care databases.

METHODS

We conducted a retrospective cohort study using MIMIC-IV v3.1 and eICU-CRD v2.0. Adult female ICU stays with gynecologic malignancy were identified by ICD codes in MIMIC-IV and diagnosis strings in eICU-CRD. The primary outcome was in-hospital mortality. The primary pooled model evaluated age, tumor group, database-recorded sepsis, database-recorded acute kidney injury, metastatic cancer, and database cohort. Additional sensitivity analyses incorporated first-day Sequential Organ Failure Assessment (SOFA) in MIMIC-IV, Acute Physiology and Chronic Health Evaluation (APACHE) IVa in eICU-CRD, first-24-hour organ support, laboratory-burden proxies, patient-level clustering, hospital/center-cluster robust standard errors, and database interaction terms.

RESULTS

The study included 711 ICU stays, including 248 from MIMIC-IV and 463 from eICU-CRD. Overall in-hospital mortality was 14.5%. In the primary pooled model, database-recorded sepsis (OR 2.17, 95% CI 1.35-3.49), database-recorded acute kidney injury (OR 2.27, 95% CI 1.38-3.71), and metastatic cancer (OR 2.03, 95% CI 1.14-3.59) were associated with higher in-hospital mortality. Severity scores were higher among nonsurvivors (median SOFA 6.0 vs. 3.0 in MIMIC-IV; median APACHE IVa 77.0 vs. 50.0 in eICU-CRD). In database-specific severity models, SOFA (OR 1.30 per point, 95% CI 1.14-1.49) and APACHE IVa (OR 1.04 per point, 95% CI 1.03-1.05) were strongly associated with mortality, while sepsis and acute kidney injury estimates attenuated.

CONCLUSIONS

Critically ill women with gynecologic malignancies had substantial in-hospital mortality. Database-recorded sepsis, database-recorded acute kidney injury, and metastatic cancer identified higher-risk patients in the primary model, but severity-adjusted analyses suggest that these variables should be interpreted mainly as descriptive markers of acute illness burden rather than independent causal effects.